Patients who suffer a myocardial infarction (MI) are at risk of late sudden cardiac death. The underlying cause can be divided into 2 groups: recurrent acute MI or cardiac rupture, and arrhythmias. 1 The second group might benefit from an implantable cardioverterdefibrillator, but identification of those patients who will benefit the most remains a continuous challenge.Recently, the 2-year follow-up of the Cardiac Arrhythmias and Risk Stratification After Acute Myocardial Infarction (CARISMA) trial was published in Circulation and commented on by Buxton. 2 The study showed that, among 297 patients with a left ventricular ejection fraction Յ40% after an acute MI, a high-degree atrioventricular (AV) block was the most powerful predictor of cardiac death, and secondarily, the heart rate variability analysis in the longer period (after 6 weeks), as shown by previously published results.The CARISMA investigators also conclude that the incidence of clinically relevant arrhythmias might have been underestimated because of limitations of the implantable loop recorder (ILR).The main messages we take from CARISMA is that we underestimate the incidence/clinical relevance of atrial and ventricular arrhythmias, and that continuous monitoring, in selected patients, may be a valuable choice if improved and adjusted.Recently, Scirica et al 3 showed that short runs of nonsustained ventricular tachycardia (Ն4 beats) still existing 48 hours after MI were associated with a higher risk of sudden cardiac death. It was not possible to identify these early arrhythmias in the CARISMA Study due to 2 factors: time of implantation and settings of the ILR. Because post-MI patients are identified by having an event, the ILR will always be implanted after the event has taken place (in CARISMA, mean of 11 days after MI), and may miss the early detection window.We would like to offer a complementary approach for a better understanding: different populations with a high risk of cardiac complications that might provide a solution for capturing these critical nonsustained ventricular tachycardia. We experienced a high incidence of new-onset arrhythmias and MI in elective vascular surgery patients. 4,5 Using 72-hour Holter recordings, we observed an incidence of perioperative arrhythmias of 11%, 21% of which were recorded after troponin release. 4 With an ILR, the true incidence might prove to be even higher.In addition, the settings of the ILR are of importance. In CARISMA, a ventricular tachycardia would only be identified if every beat was at a rate of 125 beats per minute, with a minimal length of 16 beats. In order to detect all clinically relevant arrhythmias, the settings need to have a higher specificity and programmability, which are now available in newer generations of ILR.To further improve identification of patients at risk of sudden cardiac death directly post-MI, the ILR can be combined with a Holter recording. In this way, heart rate variability and presence of nonsustained ventricular tachycardia, which are predictors of later arrh...